titleeprints.ucm.es/47934/1/sc_2011_56_4_297.pdf · particularly after adjustmentof the bead size...
TRANSCRIPT
Title:
Cleaning methods assessment for the limestone façades of the formerly Workers
Hospital of Madrid, Spain
Authors:
Elena M. Perez-Monserrata,b
Maria J. Varasa,b ([email protected])
Rafael Forta ([email protected])
Monica Alvarez de Buergoa* ([email protected])
a Instituto de Geología Económica (CSIC-UCM) [Institute of Economic Geology,
Spanish Council for Scientific Research (CSIC) and Complutense University of Madrid
(UCM)], José Antonio Novais 2, 28040 Madrid, Spain
b Department of Petrology. School of Geology, Complutense University of Madrid
(UCM), José Antonio Novais 2, 28040 Madrid, Spain
* Corresponding author (M. Alvarez). Tel.: +34 913944902; fax: +34 915442535
Abstract
Four methods for cleaning the limestone facades on what was originally the Workers' Hospital are
analyzed in this paper. Due to the pollution in the surrounding air, just twenty years after a prior
cleaning operation, sulfate crusts had developed on the entire stone surface of the building. The
gypsum mortar used in the original masonry constituted an additional source of sulfur.
Limestone is a traditional building material in Madrid and surroundings. The petrography, mineralogy
and petrophysical properties of the biomicrite, pelmicrite and biopelmicrite varieties identified in the
hospital walls were determined. An analysis of the black layer on the stone surface showed that it
consisted primarily of sulfate crusts. The cleaning methods tested were alkaline gels (sodium hydroxide
and potassium hydroxide), pressurized hot water, glass bead blasting and latex peeling. The criteria for
assessing the effectiveness and potential dangerousness of the various cleaning systems included
changes in the chromatic parameters of the clean stone, as well as formation of alteration products (i.e.,
salts) and modification of the stone surface. The stones cleaned with the three most effective methods
as well as the rain-washed stone used as a reference were washed to generate an artificial runoff.
The drain water collected was analyzed to determine possible cleaning process by-products.
The method found to be most effective and alter the stone surface the least was glass bead blasting,
particularly after adjustment of the bead size and pressure conditions used for the test.
Keywords: limestone, cleaning methods, assessment, chromatic parameters, SEM-EDS
1. Introduction
The former Workers’ Hospital, designed by architect Antonio Palacios, was built between 1909 and
1916, primarily of limestone. Also used on the surrounding wall (Figure 1), the stone is from the Caliza
del Páramo geological formation located in the east and southeast of the Community of Madrid.
Several stone varieties are found in this formation, all of which have been widely used in the built
heritage of the region since the 18th century [1,2]. The building was used as a hospital until the early
nineteen seventies, when it closed chiefly for want of financial support [3], and remained vacant for 15
years. In 1979 it was listed as a national artistic and historic monument. The regional Government of
Madrid purchased the property in 1984 as the headquarters for the regional Department of Housing and
Transport. In the comprehensive restoration and rehabilitation works performed between 1984 and
1986, one of the main tasks was to clean and protect the stone facades.
Accumulated surface deposits are responsible for the recent re-blackening of the facades, with the
concomitant loss of aesthetic value [4]. The main forms of decay observed in the hospital’s stone
facades are related to such black layers, which on much of the stone surface is associated with the
development of sulfate crusts. Where air pollution is severe, sulfate crusts tend to develop on carbonate
stones not washed by rainwater, which progressively blacken due to the deposition of particulate matter
[5-9]. While the polluted urban environment where the building stands is the primary cause of facade
soiling and blackening, there are others, including the many insets in the facade and wall design that
hinder the rainwater washing of the elements involved [10]; the rusticated finish of the ashlars that
favours particle deposition [11]; and the passing of time itself.
The facade cleaning underway since June 2006, with completion scheduled about the middle of 2008,
aims primarily to remove stone soiling and blackening and enhance building aesthetics. Prior studies to
assess the most suitable cleaning method were requested. Today’s preventive policies for historic
buildings advocate maintenance and urban remodelling to reduce the effects of atmospheric pollution
[12]. As a rule, however, it is difficult to balance air pollution and building material decay on the one
hand and the aesthetic value and maintenance of buildings on the other [13-15]. A complete description
of testing procedures to assess cleaning treatments is in [16], with a list of properties to measure, test
methods to use, evaluated parameters and criteria for positive evaluation, as well as the characteristics
of evaluation tests.
The present study aimed to identify the most appropriate method for cleaning the hospital’s stone
facades, taking into consideration the properties of the stone and their surface deposits.
2. Materials and Methods
Several studies were conducted to assess the most effective and suitable method for cleaning the stone.
Four techniques were pre-selected and subsequently tested in situ on the stone in the surrounding wall.
This area was chosen for testing because it is one of the most severely soiled parts of the building due
to its direct exposure to heavy vehicle traffic (Figure 1) and to its position, set back from the edge of
the lot (Figure 2). The surrounding environment was analyzed; the stone and the joint mortar were
characterized; the nature of the soiling was analyzed before proceeding to clean the stone; and the
advantages-drawbacks of the cleaning methods selected were studied, because building clean-up
operations are governed by both aesthetic and materials science criteria [17].
2.1. Surrounding air
Twenty years after they were last cleaned (1986), the stone facades were severely soiled, due primarily
to the building’s location in a polluted urban area (Figure 1), one of the busiest in the city where
vehicle traffic is especially heavy. The flyover that stood on the street bordering the building for 35
years (1969-2005) supported traffic on the order of 80 000 vehicles over top and 70 000 underneath per
day. The vehicle entry to the present underpass built in 2005 to replace the flyover is just opposite the
main facade of the building (Figure 1). Figure 3 shows the concentration by month of SO2 and
particulate matter from 1986 to 2006. The data were obtained from Madrid’s Municipal Air Quality
Surveillance Network. Specifically, the values were recorded at the Cuatro Caminos environmental
measurement station, which was located just 150 m from the building, although since December 1998
it was sited at a distance of 380 m from the hospital. Note that the concentration of both pollutants has
declined, particulate matter more slightly and SO2 much more sharply, particularly as regards the
highest levels. Even where air quality improves, stone has a “stored memory” of past pollution,
although not all authors agree that this is the case [18]. In any event, periodic monitoring of stone
facade soiling from this time on would be highly recommendable.
2.2. Building materials
Stone and mortar samples were studied with an Olympus BX51 polarized microscope fitted with an
Olympus DP12 (6V/2,5Å) digital camera and proprietary software. Thin sections were stained with
alizarin red-S to differentiate carbonate minerals. X-ray diffraction techniques were used to study stone
mineralogy on a Philips PW1752, 40 kV, 30 mA diffractor with a Cu source anode, º2Ө, and PC-ADP
diffraction software.
The petrophysical characterization of the limestone consisted in determining: the chromatic parameters
(L* -Luminosity-, a* and b* - chromatic coordinates, a* accounts for red-green, and b* for yellow-
blue-, C* -Chroma-, WI -White Index- and YI –Yellow Index-) [19] using a Minolta CM-2002
spectrophotometer with CIE Standard Illuminant D65 and a 10º observer angle; porosity accessible to
water or open porosity and water saturation, obtained through the vacuum saturation method; pore size
distribution (differentiating between micro and macroporosity, being 5 µm pore diameter the
threshold), and porosity accessible to mercury, with a Micromeritics AutoPore IV 9520 Hg-porosimeter
having a pore diameter range from 0.001 to 1000 µm and measuring conditions from atmospheric
pressure to 60 000 psia (228 MPa); and the water-stone contact angle with a CAHN, DCA-315
dynamic contact angle analyzer, by means of which, a value expressed in degrees: the higher the value
(>90º), the more water-repellent or hydrophobic the analyzed stone surface.
The stones found most commonly throughout the building are classified as biomicrites, pelmicrites and
biopelmicrites [20] (Figure 4), with calcite as the main and quartz as the accessory mineral. While
known locally as Colmenar stone, its petrology does not exactly conform to that variety. Despite the
heterogeneity, in some measure, of its physical properties (Table 1), this limestone is a high quality
building material. It has a low open porosity (or porosity accessible to water), although most of the
pores are in the range of micropores. Water stone contact angle measuring show that it is not a very
water-repellent or hydrophobic stone by itself, although the maximum quantity of water that the stone
is able to contain (saturation water, under vacuum in the laboratory) is not significant (1.63%±0.76).
Several layers of mortars were found in the masonry joints. The outer-most cement mortar covers what
is probably the original, gypsum mortar. The cement mortar contains siliceous aggregates (quartz,
feldspar, rock fragments, mica) in a dark, lumpy, microcrystalline matrix. The mineralogical
composition of the cement mortar includes a siliceous aggregate and an aluminium-rich cement
(gehlenite, alite, belite, tricalcium aluminate and celite). The degree of carbonation (calcite) and
sulfation (gypsum) of this kind of mortar, which is indicative of decay, varied depending on the area
chosen for sampling (Figure 5).
The gypsum mortar also contained siliceous aggregate (quartz, igneous rock fragments, feldspar, mica
and ceramic fragments or chamotte) in its composition, as well as some calcitic particles, in a dense
and fibrous gypsum matrix (Figure 6). Large gypsum crystals (up to 6 mm), the result of incomplete
mixing, were also observed (Figure 7). These crystals exhibited corrosion around the edges, an
outcome of successive dissolution and precipitation. Gypsum (CaSO4·2H2O) and bassanite
(CaSO4·½H2O) were found in their mineralogical composition.
2.3. Type of stone surface deposits
The samples were first studied under a polarized microscope to determine their nature. Where
warranted by sample thinness and lack of cohesion, the specimens were vacuum-impregnated with
resin. Scanning electron microscopy (SEM) was also utilized, using a JEOL JSM 6400 SEM with an
acceleration voltage of 2 to 40 kV, a beam current of 6 x 10-10 A, 10-5 Torr vacuum conditions, and
resolution of 35 Å; the working distance was 8 mm at 35 kV. The images were recorded at an
acceleration voltage of 20 kV. An Oxford-INCA X-ray Energy Dispersive Spectrometer (EDS) with a
nominal resolution of 133 eV at 5.39 kV was coupled to the SEM. Samples were graphite sputtered
(15-mm thick cover) with a Balzers Med010 sputter coater (Balzers, Liechtenstein).
2.4. Selection of cleaning methods
The grounds for selecting cleaning methods were: traditional use for a long time in heritage buildings,
at least in Spain, and still nowadays, unfortunately (gel-based systems); satisfactory results in other
stone buildings (glass bead blasting); or least aggressive approach (latex) [21-26].
Four groups of cleaning methods were selected:
1: alkaline gel-based system (sodium and potassium hydroxides). The testing area to be cleaned is
water sprayed, the gel applied by brushing for 30 to 45 minutes, and then the product is removed with a
water jet (at 5.88 MPa and 60 ºC).
2: pressurized hot water (at 5.88 MPa and 60 ºC)
3: wet glass bead blasting with a bead size ranging from 50 to 100 micrometers in diameter and a
water:bead ratio of 1:4, under a pressure of 5.88 MPa
4: latex peeling with a product containing 10 % EDTA (ethylenediaminetetraacetic acid) and ammonia.
The testing area must be dry and the product is applied by brushing, leaving it for 24 hours and then
removing it, directly pulling from the latex as if it were a peeling mask.
While the use of the first three groups of methods for cleaning stone is widespread and amply
documented, no results on stone cleaning with latex could be found in the literature.
2.5. Cleaning method effectiveness
The effectiveness of these cleaning methods was assessed by determining the chromatic parameters of
the limestone prior to and after cleaning, as well as the overall colour change (∆E* = (∆L*2 + ∆a*2 +
∆b*2)½) [8,22,27-29]. The reference value was taken to be the colour of the naturally rain-washed
limestone, primarily the limestone at the socle of the building, as per Fort et al., 2000c. The closer the
colour of the clean stone was to the rain-washed stone reference, i.e., the lower the ∆E* value, the more
efficient was the cleaning method.
Ca. of two hundred and forty parameters were obtained, by means of three measurements on every field
area: fresh limestone, naturally rain-washed socle and on five different areas before and after cleaning
with the five different methods.
The other criteria for assessing the suitability of the cleaning methods were the presence or absence of
alteration products (i.e. salt formation) and stone surface modifications. A Metrohm 761 Compact IC
ion chromatographer was used for this purpose, in conjunction with a chemical suppressor module N/S
1761002017199 and IC Net 2.3 software. The separation columns used were:
- a Metrosep A Supp 5 – 150, N/S 7209533 anion column, with a standard and maximum flow of
0.7 and 0.8 mL/min, respectively, and a pressure of 6.5MPa. Regarding to eluents, sodium
hydrogen carbonate (168mg/2 L-1.0 mmol/L) and sodium carbonate (678mg/2 L–3.2 mmol/L)
were used as carbonate eluent; and sodium hydroxide (1600mg/2 L-20.0 mmol/L) and p-
Cyanophenol (476mg/2 L - 4.0 mmol/L) were used as hydroxide eluent.The injection flux was
20 µl and ultrapure water was employed for rinsing.
- a Metrosep C2 150 N/S 1116.0029 cation column, with a standard and maximum flow of 1.0
and 5 mL/min, respectively, and a pressure of 7.0MPa. Regarding to eluents, tartaric acid
(1200mg/2 L-4.0 mmol/L) and dipicolinic acid (250mg/2 L–0.75 mmol/L) were used as
standard eluent; and nitric acid (c=1mol/L, 4mL/2L-2.0 mmol/L).The injection flux was 20 µl
and ultrapure water was employed for rinsing.
Both the stones cleaned with the three most effective methods and the rain-washed stone were rinsed to
generate an artificial runoff. The height of wall affected by the runoff – less than one litre of water –
was around 1 m. The water collected was analyzed to determine possible cleaning process by-products
on the stone surface. Anions determined were chloride, nitrate and sulfate; cations were sodium,
potassium and calcium.
The selection of these anions and cations was performed considering them as forming part of some of
the most frequent salts as by-products, especially taking into account the nature and composition of
materials and by-products, as well as considering the nature of the cleaning gels.
3. Results and Discussion
3.1. Type of surface deposits
The main component of the accumulated surface deposits was gypsum (Figures 4-8). There were two
layers of crust: the outer-most, not always present, was irregular and around 10-20 µm thick. It
contained gypsum, Si, Al and C, as well as the remains of micro-organisms. This layer was compact,
irregular and lumpy in appearance. The inner layer, in contact with the stone, was thicker and
composed of more intensely crystallized gypsum and halite (NaCl). In some places, this crust was up to
3 mm thick. Prismatic gypsum crystals were sometimes found in stone fissures. This is one of the
reasons that made stone cleaning necessary, for surface deposits accumulation and blackening cause
not only aesthetic damage (soiling) but may actually lead to physical and chemical decay of the stone.
The origin of this black crust may be traced to the polluted environment where the building is located;
the main component, gypsum, forms from atmospheric SO2 (high concentrations of which are in the air
due to motor vehicle emissions) as the source of sulfur, and the calcium from the calcite in the
limestone [6,30-33]. Another source of sulfur is the original gypsum mortar. A third possible source
that cannot be ruled out is the sulfur from cement mortar sulfation. The primary sources of the water
needed to dissolve the gypsum are the periodic sprinkling of the landscaped areas between the outer
wall and the main building and rainwater.
3.2. Assessment of the stone cleaning methods
3.2.1. Overall colour change (∆E*)
The sodium hydroxide-based cleaning system proved to be acceptably effective. The results of the
potassium hydroxide-based method, on the contrary, were unsatisfactory, for this product left a whitish
veil on the stone surface. Glass bead blasting was also effective, removing the crust without generating
surface by-products. Neither the latex nor the pressurized water methods were acceptable, for they led
to no change in the surface visible to the naked eye (Figure 9); moreover, not all the latex was readily
removable.
Table 2 shows that the alkaline gel- and glass bead-based cleaning methods yielded the highest
luminosity (L*) values, approaching the values recorded for the rain-washed limestone. The yellow
index (YI) values measured on the stone after cleaning with these systems were also very similar to the
reference values. The lowest ∆E* value measured after cleaning was obtained with the NaOH-based
gel, followed by the KOH-based method and glass bead blasting. The ∆E* values measured on the
limestone cleaned with the pressurized water and latex methods declined slightly from the values found
prior to cleaning. The smaller the modification of the yellow and white indices obtained (with respect
to the naturally washed stone), the better and more efficient is a cleaning system. The white index
found for the limestone tested with the NaOH alkaline gel treatment was very similar to the rain-
washed stone index. The latex and pressurized water systems were less effective from this standpoint,
for the limestone exhibited similar chromatic parameters before and after cleaning. In terms of overall
colour change, the lower the values after cleaning, the more effective is the method, because the whole
point of cleaning is to turn the colour of the soiled stone to a tone as close as possible to the colour of
the naturally washed stone.
A comparison of the white and yellow indices of the samples cleaned with the various systems to the
rain-washed reference showed that the glass bead and alkaline gel methods were the most effective in
this regard, as they yielded the smallest chromatic differences (Figure 10).
3.2.2. Cleaning by-products
The analysis of the artificial runoff collected from the stones cleaned with the three most effective
methods and from the rain-washed stone by means of ion chromatography is shown in Table 3. The
high anion content measured on the naturally washed stone may have been partly due to the
accumulation of urine and droppings [34] and the natural runoff over the above black crusts.
While the three most effective methods reduced the chloride, nitrate and sulfate content on the stone
surface, none achieved complete removal. The most efficient in this regard was glass bead blasting.
All three methods raised the content of the selected cations (Na, K, Ca). In the case of the KOH gel,
this compound would account for the potassium content and the whitish veil on the stone surface.
3.2.4. Textural changes on the stone surface
The degree of soiling removed from the limestone surface with the different cleaning methods was
determined with SEM (+EDS). Gypsum crystals were found in sheltered areas of the stone surface
cleaned with a sodium hydroxide-based gel (Figure 11a). Moreover, some of the calcite crystals in the
area where the black crust was eliminated were slightly dissolved (Figure 11b), an outcome of the
fairly aggressive nature of this product.
The KOH gel thinned the black crust but did not remove it entirely. In addition, the high potassium
content may have induced salt formation (as already suggested by [35] and with it the whitish
postcleaning veil observed on the stone surface (Figure 11b). Glass bead blasting was the most
effective method, for it removed the black crust completely, i.e., both soiling and the gypsum crystals.
In addition, bead impact left only faint marks on only a few of the calcite crystals (Figure 11c).
4. Conclusions
Of the four cleaning methods selected, pressurized water and latex peeling proved to be ineffective. In
terms of chromatic criteria (∆E*, ∆YI and ∆WI), the most effective methods were alkaline gel-based
systems and glass bead blasting. All three systems yielded clean stone whose colour was close to the
colour of the naturally washed stone used as a reference. The methods were found to have drawbacks,
however. The KOH based gel may lead to the formation of salt by-products (whitish veil) and the glass
bead system may affect the surface texture slightly. The NaOH based alkaline gel, in turn, may increase
the Na content on the surface of the stone, with the concomitant risk of a future build-up of potentially
destructive sodium sulfates. Be it said, however, that the SEM/EDS results showed no sodium remains
on the clean stone surface, at least on the samples that were analysed. Regarding the three methods to
evaluate de effectiveness of the cleaning systems (colour parameters variation, cleaning by products by
ion chromatography and textural changes on stone surface by SEM), the least reliable was the ion
chromatography method, with results difficult to interpret, probably due to the methodology followed.
Given the properties of the stone in question, namely a stone-water contact angle that is not particularly
high and a predominance of microporosity, the NaOH based alkaline gel was ruled out, for it would
entail the use of vast quantities of water and the attendant likelihood of salt remobilization; the greater
the microporosity, the higher is the potential for salt crystallization-mediated decay.
Consequently, the conclusion drawn from the results of this study is that the most effective method for
cleaning the limestone on this building is glass bead blasting.
Nonetheless, a smaller bead size and lower working pressure are recommended to prevent damage to
the stone surface. And indeed, this is the method that was used to clean the facades, with quite
satisfactory results and no damage to the stone.
In light of the building location – in an area with heavy traffic –, black crusts will definitely develop
and soiling will accumulate in the not-so-distant future. The recommendation, therefore, is for facade
cleaning to be included as a part of building maintenance. On the one hand, facade maintenance should
take place more frequently and involve simpler cleaning operations, rather than huge and expensive
repair works.
And on the other, preventive maintenance, including in situ measurements and regular monitoring of
relevant parameters to estimate the rate of facade soiling and blackening.
Final note from the authors: This case study represents a real example of how scientific method can
contribute to optimal restoration and conservation of the architectural heritage. While it may appear to
be obvious that some of the cleaning methods discussed should not have even been tested,
unfortunately in the real world, some of these systems continue to be used indiscriminately despite the
lack of any scientific endorsement. Even today, only a small number of institutions responsible for
heritage conservation are aware of the need to conduct scientific studies prior to any manner of
intervention. In this specific case, however, the Community of Madrid did undertake preliminary
testing to ascertain the effectiveness of the cleaning methods under consideration for the building
studied here. The outcome was that the methods that proved to be ineffective or damaging to the stone
to be cleaned were ruled out. While this may appear to be obvious, scientific evidence was required to
select the optimal method in terms of both efficacy and harmlessness for the stone. This type of
collaboration aims to encourage the institutions responsible to trust in science to solve what may appear
to be merely technical problems. A second objective is to generate the need for studies that contribute
to the conservation of the cultural heritage and at the same time and no less importantly to medium-
and long-term savings on the costs of such interventions.
Acknowledgments
This work has been carried out as part of a Special Intramural Project, supported by the Spanish
National Research Council. We would like to express our gratitude to José María La Calle, architect of
the Architecture and Rehabilitation Directorate (Regional Housing Ministry), and to PROLISER and
FTB companies, both specialized in façades cleaning, the facilities and assistance provided.
This study was funded by GEOMATERIALES (S2009/MAT-1629) and CONSOLIDER-TCP
(CSD2007-0058). We thank the UCM Research Group ‘‘Alteracion y Conservacion de los Materiales
Petreos del Patrimonio’’ (ref. 921349).
Finally, we thank Margaret Clark for the English proofreading of the manuscript.
References
1. Dapena, J.E., Ordóñez S., and García del Cura, M.A., ‘Estudio de las rocas calizas utilizadas
durante los siglos XVIII y XIX en la construcción de los palacios de Madrid’, Ingeniería Civil 71 (1989) 71-77.
2. Fort, R., Mingarro, F., and López de Azcona, M.C., ‘Petrología de los materiales de
construcción del Palacio Real de Madrid’, Geogaceta 20(5) (1996) 1236-1239 3. Ciudad, A., El hospital de Maudes: la adaptación de un edificio a través de la historia,
Consejería de Obras Públicas, Urbanismo y Transportes de la Comunidad de Madrid (ed.),
Madrid (2001).
4. Grossi, C.M., and Brimblecombe, P., ‘Effect of long-term changes in air pollution and climate
on the decay and blackening of European stone buildings’, ed. R. Prikryl and B.J. Smith,
Geological Society, Special Publication 271 (2007) 119-130.
5. Saiz-Jimenez, C., ’Deposition of airborne organic pollutants on historic buildings’, Atmospheric Environment 27B (1993) 77-85.
6. Ausset, P., Del Monte, M., and Lefevre, R.A. ’Embryonic sulphated black crusts on carbonate
rocks in atmospheric simulation chamber and in the field: role of carbonaceous fly-ash’, Atmospheric Environment 33 (1999) 1525-1534.
7. Esbert, R.M., Díaz-Pache, F., Grossi, C.M., Alonso, F.J., and Ordaz, J., ’Airborne particulate matter around the Cathedral of Burgos (Castilla y León, Spain)’, Atmospheric Environment 35 (2001) 441-452.
8. Grossi, C.M., Esbert, R.M., Díaz-Pache, F., and Alonso, F.J., ’Soiling of building stones in urban environments’, Building and Environment 38 (2003) 147-159.
9. Bonazza, A., Sabbioni, C., and Ghedini, N., ’Quantitative data on carbon fractions in interpretation of black crusts and soling on European built heritage’, Atmospheric Environment 39 (2005) 2607-2618.
10. Pospísil, S., Drdácky, M., Slizková, Z., Knotková, D., ’Surface degradation of complex
architectural form due to atmospheric pollution’, in Heritage, Weathering and Conservation, ed. R. Fort, M. Alvarez de Buergo, M. Gomez-Heras and M.C. Vazquez-Calvo, Balkema, Leiden,
vol 1, (2006) 447-452.
11. Benavente, D., Martínez-Verdú, F., Bernabeu, A., Vigueira, V., Fort, R., García del Cura, M.A.,
Illueca, C., and Ordóñez, S., ’Influence of surface roughness on colour changes in building stones’, Colour Research Application 28 (2003) 343-351.
12. Fort, R., Alvarez de Buergo, M., and López de Azcona, M.C., ’The efficiency of urban remodelling in reducing the effects of atmospheric pollution on monuments’, in Air Pollution and Cultural Heritage, ed. C. Saiz-Jimenez, Balkema, Amsterdam, (2004) 225-232.
13. Laurenzi, M., ’Air pollution and maintenance of buildings: an unpleasant, costly correlation’, in Patrimonio Histórico Artístico y Contaminación: Encuentro Europeo, ed. L. Carbó et al.,
Madrid, Consorcio para la organización de Madrid capital Europea de la cultura (1992) 101-
105.
14. Ball, J., Laing, R.A., and Young, M.E., ’Stone cleaning: comparing perceptions with physical
and financial implications’, Journal of Architectural Conservation 6 (2000) 47–62. 15. Young, M.E., Urquhart, D.C.M., and Laing, R.A., ’Maintenance and repair for stone cleaned
sandstone and granite building façades’, Building and Environment 38 (2003) 1125-1131. 16. Tabasso, M.L., and Simon, S., ’Testing methods and criteria for the selection/evaluation of
products for the conservation of porous building materials’, Reviews in Conservation 7 (2006) 67-82.
17. Kavenagh, C., and Wheeler, G., ’Evaluation of cleaning methods for the exterior brick at the
Brooklyn Historical Society’, Journal of the American Institute for Conservation 42 (2003) 97-112.
18. Vleugels, G, Dewolfs, R, and Van Grieken R., ’On the memory effect of limestone for air
pollution’, Atmospheric Environment 27 (1993) 1931-1934. 19. CIELAB. CIE—Commission Internationale de l'Eclairage, ’Recommendations on uniform
colour spaces, colour difference equations and psychometric colour terms’. Colourimetry, Supplement (1978) 2–15.
20. Folk, R.L., ’Spectral subdivision of limestone types’, AAPG Mem. 1 (1962) 62–84. 21. Fort, R., López de Azcona, M.C., Mingarro, F., Alvarez de Buergo, M., and Rodríguez J.,
’Assessment of protective treatments based on their chromatic evolution: limestone and granite
in the Royal Palace of Madrid, Spain’, in Protection and Conservation of the Cultural Heritage of Mediterranean Cities, ed. E. Galán, F. Zezza, Balkema, Lisse, (2000) 437-441.
22. Fort, R., López de Azcona, M.C., Mingarro, F., and Rodríguez, J., ’Chromatic parameters as
performances indicators for stone cleaning techniques’, COLOR Research and Application 25(6) (2000) 442-446.
23. Fort, R., López de Azcona, M.C., and Mingarro, F., ’Cleaning of stone materials in the
Cathedral of Valladolid (Spain)’, Materiales de Construcción 50(258) (2000) 37-49. 24. Moropoulou, A., and Kefalonitou S., ’Efficency and countereffects of cleaning treatment on
limestones surfaces-investigation on the Corfu Venetian Fortress’, Building and Environment 37 (2002) 1181-1191.
25. Iglesias, M., Gea, B., Prada, J.L., and Guasch, N., ’Low-pressure abrasive cleaning of historic buildings materials’, in Heritage, Weathering and Conservation, ed. R. Fort, M. Alvarez de
Buergo, M. Gomez-Heras and M.C. Vazquez-Calvo, vol 2, Balkema, Leiden, (2006) 681-686.
26. Varas, M.J., Alvarez de Buergo, M., and Fort, R., ’The influence of past protective treatments
on the deterioration of historic stone façades’, Studies in Conservation 52 (2007) 110-124. 27. Boutin, F., and Leroux, L., ’Colour and weight evolution of limestones protected by water
repellents after three-year ageing period in urban conditions’, in 9th International Congress on
Deterioration and Conservation of Stone, ed. V. Fassina, vol 1, Elsevier, Amsterdam (2000)
197-205.
28. Grossi, C.M., Esbert, R.M., Alonso, F.J., Valdeón, L., Ordaz, J., and Díaz-Pache F., ’Colour changes and reactivity to SO2 of some cladding stones at the Gran Theater del Liceu
(Barcelona-Spain)’, in 9th International Congress on Deterioration and Conservation of Stone, ed. V. Fassina, vol 1, Elsevier, Amsterdam (2000) 323-328.
29. Grossi, C.M., Brimblecombe, P., Esbert, R.M., and Alonso, F.J., ’Colour changes in architectural limestones from pollution and cleaning’, Colour Research and Application 32(4) (2007) 320-331.
30. Moropoulou, A., Bisbikoua, K., Torfsa, K., Van Grieken, R., Zezza, F., and Macri F., ’Origin and growth of weathering crusts on ancient marbles in industrial atmosphere’, Atmospheric Environment 32 (1998) 967-982.
31. Rodriguez-Navarro, C., and Sebastian, E., ’Role of particulate matter from vehicle exhaust on
porous building stones (limestone) sulfation’, Science of The Total Environment 187 (1996) 79-91.
32. Maravelaki-Kalaitzaki, P., and Biscontin, G., ’Origin, characteristics and morphology of
weathering crusts on Istria stone in Venice’, Atmospheric Environment 33 (1999) 1699-1709. 33. Bugini, R., Tabasso, M.L., and Realini M., ’Rate of formation of black crusts on marble. A case
study’, Journal of Cultural Heritage 1 (2000) 111-116. 34. Gomez-Heras, M., Benavente, D., Fort, R., and Alvarez de Buergo, M., ’A note on soluble salt
minerals from pigeon droppings as potential contributors to the decay of stone base cultural
heritage’, European Journal of Mineralogy 16 (2004) 505-509. 35. Tsui, N., Flatt, R.J., and Scherer, G.W., ’Crystallization damage by sodium sulphate’, Journal of
Cultural Heritage 4 (2003) 109-115.
Porosity accessible to water (%) 4.1 ± 1.8
Water saturation (%) 1.6 ± 0.8
Porosity accessible to Hg (%) 2.5 ± 1.7
Macroporosity >5 µm (%) 23.5 ± 25.6
Microporosity <5 µm (%) 76.5 ± 25.6
Average pore size (µm) 0.05 ± 0.02
Water-stone contact angle (º) 61 ± 5
Chromatic parameters
L*
a*
b*
C*
WI
YI
75 ± 2
2 ± 1
6 ± 1
7 ± 1
27 ± 6
11 ± 2
Table 1. Petrophysical properties of the limestone
L* a* b* C YI WI
NaOH BC 51.2±9.5 1.9±0.1 7.5±1.0 7.7±1.0 18.4±0.2 5.4±2.3 AC 72.4±1.0 1.4±0.5 7.0±0.6 7.2±.0.7 13.4±1.3 20.8±2.9 KOH BC 42.3±1.3 3.6±0.3 12.6±0.5 13.1±0.6 33.6±1.6 -4.4±0.6 AC 69.1±3.6 2.3±1.6 7.9±1.0 8.3±1.1 15.7±2.4 15.1±5.8 Glass beads BC 47.5±2.6 2.4±07 9.6±1.1 9.9±1.2 24.6±3.5 0.4±2.4 AC 66.9±3.4 1.9±0.2 8.2±0.8 8.4±0.8 16.5±1.2 12.5±2.1 Water BC 46.2±4.4 3.0±0.5 11.4±1.1 11.8±1.2 29.2±3.4 -2.5±1.8 AC 49.9±2.1 2.5±0.9 9.1±1.5 9.4±1.7 22.6±3.9 1.9±2.9 Latex BC 44.8±0.8 2.8±0.3 9.5±0.5 9.9±0.6 25.3±1.0 -0.2±0.6 AC 52.9±2.3 2.2±0.5 11.5±1.3 11.7±1.2 26.6±1.9 -1.5±1.7 Rain-washed stone
(reference colour) 80.6±1.9 1.2±0.1 9.1±0.8 9.2±0.8 21.8±1.5 15.6±1.0
∆L* ∆a* ∆b* ∆E* ∆YI ∆WI
NaOH Before cleaning -29.4 0.7 -1.6 29.4 2.8 -16.4
After cleaning -8.2 0.2 -2.1 8.4 -2.2 -1.0
KOH Before cleaning -38.2 2.4 3.5 38.5 18.0 -26.3
After cleaning -11.5 1.1 -1.1 11.6 0.1 -6.7
Glass beads Before cleaning -33.0 1.2 0.6 33.1 9.0 -21.4
After cleaning -13.7 0.7 -0.9 13.7 0.9 -9.4
Water Before cleaning -34.4 1.8 2.3 34.5 13.6 -24.3
After cleaning -30.7 1.3 1.0 30.7 7.0 -19.9
Latex Before cleaning -35.7 1.6 0.4 35.8 9.7 -22.0
After cleaning -27.7 1.0 2.4 27.8 11.0 -23.3
Table 2. Chromatic parameters measured on the limestone prior to and after cleaning
with the different methods: first table corresponds to the absolute values; in the second
table the variations (increase/decrease) of the chromatic parameters before and after
stone cleaning compared to rain-washed stone (reference colour) are shown.
a Source: Canal de Isabel II. Madrid waterworks utility. Minimum and maximum values for 2005
ANIONS CATIONS
Chloride Nitrate Sulphate Sodium Potassium Calcium
Drinking water a 10.0-20.0 <0.9-7.8 8.0-30.0 4.0-20.0 <2.0 5.8-34.8
Rain-washed stone 46.6 119.1 101.8 12.6 4.9 8.6
NaOH 22.5 70.7 50.4 26.6 5.4 21.4
KOH 29.4 59.4 65.9 25.9 26.3 16.6
Glass beads 17.1 1.9 35.6 23.5 3.9 13.7
Table 3. Ionic composition of artificial rain water runoff over naturally washed stone and
limestone samples cleaned with the three most effective methods, compared to domestic
cold water (mg/l).
Figure captions
Figure 1. General view of the Formerly Hospital. Façades of the northwest body, church
and surrounding wall.
Figure 2. Differential soiling of the surrounding wall.
Figure 3. Monthly values of SO2 and particulate matter concentration from 1986 to
2006. Source: Air Quality Surveillance Network, Madrid City Council.
Figure 4. Thin section (cross section) of the soiling on the stone. A and B are the two
layers of the soiling crust. C corresponds to the stone.
Figure 5. XRD spectrum from a sample of cement mortar.
Figure 6. Thin section micrograph of a gypsum mortar sample. Detail of the gypsum
rich matrix, with arrow-head shaped gypsum crystals.
Figure 7. Detail of the dissolution-precipitation borders of a large gypsum crystal.
Figure 8. SEM images (cross section, SE mode) showing A: surface soiling layer, B:
gypsum crust in contact with stone, and, C: limestone
Figure 9. Wall stones appearance after (above) and prior (below) to the testing with the
different cleaning methods.
Figure 10. White and yellow indices increase of the cleaned stones with respect to the
rainwashed limestone.
Figure 11. SEM photomicrographs, in secondary electrons mode), and the
corresponding EDS area analysis on the cleaned stone for the three more effective
cleaning methods tested. (a, b) NaOH; (c) KOH; (d) Microspheres. All of the images are
cross sections of the cleaned stone (stone surface on the top of the image), except b),
which corresponds to a view of the stone surface.
Figure 1.
Figure 2
Figure 3
0
20
40
60
80
100
120
140
160
180
200
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
µg/
m3
PART
SO2
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9.
Figure 10.
Figure 11.